import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
-import jalview.io.gff.SequenceOntologyFactory;
+import jalview.datamodel.features.SequenceFeatures;
import jalview.io.gff.SequenceOntologyI;
import jalview.schemes.ResidueProperties;
import jalview.util.Comparison;
import jalview.util.DBRefUtils;
+import jalview.util.IntRangeComparator;
import jalview.util.MapList;
import jalview.util.MappingUtils;
-import jalview.util.RangeComparator;
import jalview.util.StringUtils;
import java.io.UnsupportedEncodingException;
import java.util.Arrays;
import java.util.Collection;
import java.util.Collections;
-import java.util.Comparator;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
int toOffset = alignTo.getStart() - 1;
int sourceGapMappedLength = 0;
boolean inExon = false;
- final char[] thisSeq = alignTo.getSequence();
- final char[] thatAligned = alignFrom.getSequence();
- StringBuilder thisAligned = new StringBuilder(2 * thisSeq.length);
+ final int toLength = alignTo.getLength();
+ final int fromLength = alignFrom.getLength();
+ StringBuilder thisAligned = new StringBuilder(2 * toLength);
/*
* Traverse the 'model' aligned sequence
*/
- for (char sourceChar : thatAligned)
+ for (int i = 0; i < fromLength; i++)
{
+ char sourceChar = alignFrom.getCharAt(i);
if (sourceChar == sourceGap)
{
sourceGapMappedLength += ratio;
*/
int intronLength = 0;
while (basesWritten + toOffset < mappedCodonEnd
- && thisSeqPos < thisSeq.length)
+ && thisSeqPos < toLength)
{
- final char c = thisSeq[thisSeqPos++];
+ final char c = alignTo.getCharAt(thisSeqPos++);
if (c != myGapChar)
{
basesWritten++;
int gapsToAdd = calculateGapsToInsert(preserveMappedGaps,
preserveUnmappedGaps, sourceGapMappedLength, inExon,
trailingCopiedGap.length(), intronLength, startOfCodon);
- for (int i = 0; i < gapsToAdd; i++)
+ for (int k = 0; k < gapsToAdd; k++)
{
thisAligned.append(myGapChar);
}
* At end of model aligned sequence. Copy any remaining target sequence, optionally
* including (intron) gaps.
*/
- while (thisSeqPos < thisSeq.length)
+ while (thisSeqPos < toLength)
{
- final char c = thisSeq[thisSeqPos++];
+ final char c = alignTo.getCharAt(thisSeqPos++);
if (c != myGapChar || preserveUnmappedGaps)
{
thisAligned.append(c);
SequenceI peptide = mapping.findAlignedSequence(cdsSeq, protein);
if (peptide != null)
{
- int peptideLength = peptide.getLength();
+ final int peptideLength = peptide.getLength();
Mapping map = mapping.getMappingBetween(cdsSeq, peptide);
if (map != null)
{
{
mapList = mapList.getInverse();
}
- int cdsLength = cdsDss.getLength();
- int mappedFromLength = MappingUtils
- .getLength(mapList.getFromRanges());
+ final int cdsLength = cdsDss.getLength();
+ int mappedFromLength = MappingUtils.getLength(mapList
+ .getFromRanges());
int mappedToLength = MappingUtils
.getLength(mapList.getToRanges());
boolean addStopCodon = (cdsLength == mappedFromLength
* walk over the aligned peptide sequence and insert mapped
* codons for residues in the aligned cds sequence
*/
- char[] alignedPeptide = peptide.getSequence();
- char[] nucleotides = cdsDss.getSequence();
int copiedBases = 0;
int cdsStart = cdsDss.getStart();
int proteinPos = peptide.getStart() - 1;
int cdsCol = 0;
- for (char residue : alignedPeptide)
+
+ for (int col = 0; col < peptideLength; col++)
{
+ char residue = peptide.getCharAt(col);
+
if (Comparison.isGap(residue))
{
cdsCol += CODON_LENGTH;
{
for (int j = codon[0]; j <= codon[1]; j++)
{
- char mappedBase = nucleotides[j - cdsStart];
+ char mappedBase = cdsDss.getCharAt(j - cdsStart);
alignedCds[cdsCol++] = mappedBase;
copiedBases++;
}
* append stop codon if not mapped from protein,
* closing it up to the end of the mapped sequence
*/
- if (copiedBases == nucleotides.length - CODON_LENGTH)
+ if (copiedBases == cdsLength - CODON_LENGTH)
{
for (int i = alignedCds.length - 1; i >= 0; i--)
{
break;
}
}
- for (int i = nucleotides.length
- - CODON_LENGTH; i < nucleotides.length; i++)
+ for (int i = cdsLength - CODON_LENGTH; i < cdsLength; i++)
{
- alignedCds[cdsCol++] = nucleotides[i];
+ alignedCds[cdsCol++] = cdsDss.getCharAt(i);
}
}
cdsSeq.setSequence(new String(alignedCds));
List<SequenceI> unmappedProtein)
{
/*
- * Prefill aligned sequences with gaps before inserting aligned protein
- * residues.
+ * prefill peptide sequences with gaps
*/
int alignedWidth = alignedCodons.size();
char[] gaps = new char[alignedWidth];
Arrays.fill(gaps, protein.getGapCharacter());
- String allGaps = String.valueOf(gaps);
+ Map<SequenceI, char[]> peptides = new HashMap<>();
for (SequenceI seq : protein.getSequences())
{
if (!unmappedProtein.contains(seq))
{
- seq.setSequence(allGaps);
+ peptides.put(seq, Arrays.copyOf(gaps, gaps.length));
}
}
+ /*
+ * Traverse the codons left to right (as defined by CodonComparator)
+ * and insert peptides in each column where the sequence is mapped.
+ * This gives a peptide 'alignment' where residues are aligned if their
+ * corresponding codons occupy the same columns in the cdna alignment.
+ */
int column = 0;
for (AlignedCodon codon : alignedCodons.keySet())
{
.get(codon);
for (Entry<SequenceI, AlignedCodon> entry : columnResidues.entrySet())
{
- // place translated codon at its column position in sequence
- entry.getKey().getSequence()[column] = entry.getValue().product
- .charAt(0);
+ char residue = entry.getValue().product.charAt(0);
+ peptides.get(entry.getKey())[column] = residue;
}
column++;
}
+
+ /*
+ * and finally set the constructed sequences
+ */
+ for (Entry<SequenceI, char[]> entry : peptides.entrySet())
+ {
+ entry.getKey().setSequence(new String(entry.getValue()));
+ }
+
return 0;
}
*
* @param fromSeq
* @param toSeq
+ * @param mapping
+ * the mapping from 'fromSeq' to 'toSeq'
* @param select
* if not null, only features of this type are copied (including
* subtypes in the Sequence Ontology)
- * @param mapping
- * the mapping from 'fromSeq' to 'toSeq'
* @param omitting
*/
public static int transferFeatures(SequenceI fromSeq, SequenceI toSeq,
copyTo = copyTo.getDatasetSequence();
}
- SequenceOntologyI so = SequenceOntologyFactory.getInstance();
+ /*
+ * get features, optionally restricted by an ontology term
+ */
+ List<SequenceFeature> sfs = select == null ? fromSeq.getFeatures()
+ .getPositionalFeatures() : fromSeq.getFeatures()
+ .getFeaturesByOntology(select);
+
int count = 0;
- SequenceFeature[] sfs = fromSeq.getSequenceFeatures();
- if (sfs != null)
+ for (SequenceFeature sf : sfs)
{
- for (SequenceFeature sf : sfs)
+ String type = sf.getType();
+ boolean omit = false;
+ for (String toOmit : omitting)
{
- String type = sf.getType();
- if (select != null && !so.isA(type, select))
+ if (type.equals(toOmit))
{
- continue;
- }
- boolean omit = false;
- for (String toOmit : omitting)
- {
- if (type.equals(toOmit))
- {
- omit = true;
- }
- }
- if (omit)
- {
- continue;
+ omit = true;
}
+ }
+ if (omit)
+ {
+ continue;
+ }
- /*
- * locate the mapped range - null if either start or end is
- * not mapped (no partial overlaps are calculated)
- */
- int start = sf.getBegin();
- int end = sf.getEnd();
- int[] mappedTo = mapping.locateInTo(start, end);
- /*
- * if whole exon range doesn't map, try interpreting it
- * as 5' or 3' exon overlapping the CDS range
- */
- if (mappedTo == null)
- {
- mappedTo = mapping.locateInTo(end, end);
- if (mappedTo != null)
- {
- /*
- * end of exon is in CDS range - 5' overlap
- * to a range from the start of the peptide
- */
- mappedTo[0] = 1;
- }
- }
- if (mappedTo == null)
+ /*
+ * locate the mapped range - null if either start or end is
+ * not mapped (no partial overlaps are calculated)
+ */
+ int start = sf.getBegin();
+ int end = sf.getEnd();
+ int[] mappedTo = mapping.locateInTo(start, end);
+ /*
+ * if whole exon range doesn't map, try interpreting it
+ * as 5' or 3' exon overlapping the CDS range
+ */
+ if (mappedTo == null)
+ {
+ mappedTo = mapping.locateInTo(end, end);
+ if (mappedTo != null)
{
- mappedTo = mapping.locateInTo(start, start);
- if (mappedTo != null)
- {
- /*
- * start of exon is in CDS range - 3' overlap
- * to a range up to the end of the peptide
- */
- mappedTo[1] = toSeq.getLength();
- }
+ /*
+ * end of exon is in CDS range - 5' overlap
+ * to a range from the start of the peptide
+ */
+ mappedTo[0] = 1;
}
+ }
+ if (mappedTo == null)
+ {
+ mappedTo = mapping.locateInTo(start, start);
if (mappedTo != null)
{
- SequenceFeature copy = new SequenceFeature(sf);
- copy.setBegin(Math.min(mappedTo[0], mappedTo[1]));
- copy.setEnd(Math.max(mappedTo[0], mappedTo[1]));
- copyTo.addSequenceFeature(copy);
- count++;
+ /*
+ * start of exon is in CDS range - 3' overlap
+ * to a range up to the end of the peptide
+ */
+ mappedTo[1] = toSeq.getLength();
}
}
+ if (mappedTo != null)
+ {
+ int newBegin = Math.min(mappedTo[0], mappedTo[1]);
+ int newEnd = Math.max(mappedTo[0], mappedTo[1]);
+ SequenceFeature copy = new SequenceFeature(sf, newBegin, newEnd,
+ sf.getFeatureGroup(), sf.getScore());
+ copyTo.addSequenceFeature(copy);
+ count++;
+ }
}
return count;
}
/**
* Returns a mapping from dna to protein by inspecting sequence features of
- * type "CDS" on the dna.
+ * type "CDS" on the dna. A mapping is constructed if the total CDS feature
+ * length is 3 times the peptide length (optionally after dropping a trailing
+ * stop codon). This method does not check whether the CDS nucleotide sequence
+ * translates to the peptide sequence.
*
* @param dnaSeq
* @param proteinSeq
List<int[]> ranges = findCdsPositions(dnaSeq);
int mappedDnaLength = MappingUtils.getLength(ranges);
+ /*
+ * if not a whole number of codons, something is wrong,
+ * abort mapping
+ */
+ if (mappedDnaLength % CODON_LENGTH > 0)
+ {
+ return null;
+ }
+
int proteinLength = proteinSeq.getLength();
int proteinStart = proteinSeq.getStart();
int proteinEnd = proteinSeq.getEnd();
if (codesForResidues == (proteinLength + 1))
{
// assuming extra codon is for STOP and not in peptide
+ // todo: check trailing codon is indeed a STOP codon
codesForResidues--;
+ mappedDnaLength -= CODON_LENGTH;
+ MappingUtils.removeEndPositions(CODON_LENGTH, ranges);
}
+
if (codesForResidues == proteinLength)
{
proteinRange.add(new int[] { proteinStart, proteinEnd });
/**
* Returns a list of CDS ranges found (as sequence positions base 1), i.e. of
- * start/end positions of sequence features of type "CDS" (or a sub-type of
+ * [start, end] positions of sequence features of type "CDS" (or a sub-type of
* CDS in the Sequence Ontology). The ranges are sorted into ascending start
* position order, so this method is only valid for linear CDS in the same
* sense as the protein product.
public static List<int[]> findCdsPositions(SequenceI dnaSeq)
{
List<int[]> result = new ArrayList<int[]>();
- SequenceFeature[] sfs = dnaSeq.getSequenceFeatures();
- if (sfs == null)
+
+ List<SequenceFeature> sfs = dnaSeq.getFeatures().getFeaturesByOntology(
+ SequenceOntologyI.CDS);
+ if (sfs.isEmpty())
{
return result;
}
-
- SequenceOntologyI so = SequenceOntologyFactory.getInstance();
- int startPhase = 0;
+ SequenceFeatures.sortFeatures(sfs, true);
for (SequenceFeature sf : sfs)
{
+ int phase = 0;
+ try
+ {
+ phase = Integer.parseInt(sf.getPhase());
+ } catch (NumberFormatException e)
+ {
+ // ignore
+ }
/*
- * process a CDS feature (or a sub-type of CDS)
+ * phase > 0 on first codon means 5' incomplete - skip to the start
+ * of the next codon; example ENST00000496384
*/
- if (so.isA(sf.getType(), SequenceOntologyI.CDS))
+ int begin = sf.getBegin();
+ int end = sf.getEnd();
+ if (result.isEmpty() && phase > 0)
{
- int phase = 0;
- try
- {
- phase = Integer.parseInt(sf.getPhase());
- } catch (NumberFormatException e)
+ begin += phase;
+ if (begin > end)
{
- // ignore
+ // shouldn't happen!
+ System.err
+ .println("Error: start phase extends beyond start CDS in "
+ + dnaSeq.getName());
}
- /*
- * phase > 0 on first codon means 5' incomplete - skip to the start
- * of the next codon; example ENST00000496384
- */
- int begin = sf.getBegin();
- int end = sf.getEnd();
- if (result.isEmpty())
- {
- begin += phase;
- if (begin > end)
- {
- // shouldn't happen!
- System.err.println(
- "Error: start phase extends beyond start CDS in "
- + dnaSeq.getName());
- }
- }
- result.add(new int[] { begin, end });
}
- }
-
- /*
- * remove 'startPhase' positions (usually 0) from the first range
- * so we begin at the start of a complete codon
- */
- if (!result.isEmpty())
- {
- // TODO JAL-2022 correctly model start phase > 0
- result.get(0)[0] += startPhase;
+ result.add(new int[] { begin, end });
}
/*
* ranges are assembled in order. Other cases should not use this method,
* but instead construct an explicit mapping for CDS (e.g. EMBL parsing).
*/
- Collections.sort(result, new RangeComparator(true));
+ Collections.sort(result, IntRangeComparator.ASCENDING);
return result;
}
count += computePeptideVariants(peptide, peptidePos, codonVariants);
}
- /*
- * sort to get sequence features in start position order
- * - would be better to store in Sequence as a TreeSet or NCList?
- */
- if (peptide.getSequenceFeatures() != null)
- {
- Arrays.sort(peptide.getSequenceFeatures(),
- new Comparator<SequenceFeature>()
- {
- @Override
- public int compare(SequenceFeature o1, SequenceFeature o2)
- {
- int c = Integer.compare(o1.getBegin(), o2.getBegin());
- return c == 0 ? Integer.compare(o1.getEnd(), o2.getEnd())
- : c;
- }
- });
- }
return count;
}
String trans3Char = StringUtils
.toSentenceCase(ResidueProperties.aa2Triplet.get(trans));
String desc = "p." + residue3Char + peptidePos + trans3Char;
- // set score to 0f so 'graduated colour' option is offered! JAL-2060
SequenceFeature sf = new SequenceFeature(
SequenceOntologyI.SEQUENCE_VARIANT, desc, peptidePos,
- peptidePos, 0f, var.getSource());
+ peptidePos, var.getSource());
StringBuilder attributes = new StringBuilder(32);
String id = (String) var.variant.getValue(ID);
if (id != null)
* LinkedHashMap ensures we keep the peptide features in sequence order
*/
LinkedHashMap<Integer, List<DnaVariant>[]> variants = new LinkedHashMap<Integer, List<DnaVariant>[]>();
- SequenceOntologyI so = SequenceOntologyFactory.getInstance();
- SequenceFeature[] dnaFeatures = dnaSeq.getSequenceFeatures();
- if (dnaFeatures == null)
+ List<SequenceFeature> dnaFeatures = dnaSeq.getFeatures()
+ .getFeaturesByOntology(SequenceOntologyI.SEQUENCE_VARIANT);
+ if (dnaFeatures.isEmpty())
{
return variants;
}
// not handling multi-locus variant features
continue;
}
- if (so.isA(sf.getType(), SequenceOntologyI.SEQUENCE_VARIANT))
+ int[] mapsTo = dnaToProtein.locateInTo(dnaCol, dnaCol);
+ if (mapsTo == null)
{
- int[] mapsTo = dnaToProtein.locateInTo(dnaCol, dnaCol);
- if (mapsTo == null)
- {
- // feature doesn't lie within coding region
- continue;
- }
- int peptidePosition = mapsTo[0];
- List<DnaVariant>[] codonVariants = variants.get(peptidePosition);
- if (codonVariants == null)
- {
- codonVariants = new ArrayList[CODON_LENGTH];
- codonVariants[0] = new ArrayList<DnaVariant>();
- codonVariants[1] = new ArrayList<DnaVariant>();
- codonVariants[2] = new ArrayList<DnaVariant>();
- variants.put(peptidePosition, codonVariants);
- }
+ // feature doesn't lie within coding region
+ continue;
+ }
+ int peptidePosition = mapsTo[0];
+ List<DnaVariant>[] codonVariants = variants.get(peptidePosition);
+ if (codonVariants == null)
+ {
+ codonVariants = new ArrayList[CODON_LENGTH];
+ codonVariants[0] = new ArrayList<DnaVariant>();
+ codonVariants[1] = new ArrayList<DnaVariant>();
+ codonVariants[2] = new ArrayList<DnaVariant>();
+ variants.put(peptidePosition, codonVariants);
+ }
- /*
- * extract dna variants to a string array
- */
- String alls = (String) sf.getValue("alleles");
- if (alls == null)
- {
- continue;
- }
- String[] alleles = alls.toUpperCase().split(",");
- int i = 0;
- for (String allele : alleles)
- {
- alleles[i++] = allele.trim(); // lose any space characters "A, G"
- }
+ /*
+ * extract dna variants to a string array
+ */
+ String alls = (String) sf.getValue("alleles");
+ if (alls == null)
+ {
+ continue;
+ }
+ String[] alleles = alls.toUpperCase().split(",");
+ int i = 0;
+ for (String allele : alleles)
+ {
+ alleles[i++] = allele.trim(); // lose any space characters "A, G"
+ }
- /*
- * get this peptide's codon positions e.g. [3, 4, 5] or [4, 7, 10]
- */
- int[] codon = peptidePosition == lastPeptidePostion ? lastCodon
- : MappingUtils.flattenRanges(dnaToProtein
- .locateInFrom(peptidePosition, peptidePosition));
- lastPeptidePostion = peptidePosition;
- lastCodon = codon;
+ /*
+ * get this peptide's codon positions e.g. [3, 4, 5] or [4, 7, 10]
+ */
+ int[] codon = peptidePosition == lastPeptidePostion ? lastCodon
+ : MappingUtils.flattenRanges(dnaToProtein.locateInFrom(
+ peptidePosition, peptidePosition));
+ lastPeptidePostion = peptidePosition;
+ lastCodon = codon;
- /*
- * save nucleotide (and any variant) for each codon position
- */
- for (int codonPos = 0; codonPos < CODON_LENGTH; codonPos++)
+ /*
+ * save nucleotide (and any variant) for each codon position
+ */
+ for (int codonPos = 0; codonPos < CODON_LENGTH; codonPos++)
+ {
+ String nucleotide = String.valueOf(
+ dnaSeq.getCharAt(codon[codonPos] - dnaStart)).toUpperCase();
+ List<DnaVariant> codonVariant = codonVariants[codonPos];
+ if (codon[codonPos] == dnaCol)
{
- String nucleotide = String
- .valueOf(dnaSeq.getCharAt(codon[codonPos] - dnaStart))
- .toUpperCase();
- List<DnaVariant> codonVariant = codonVariants[codonPos];
- if (codon[codonPos] == dnaCol)
+ if (!codonVariant.isEmpty()
+ && codonVariant.get(0).variant == null)
{
- if (!codonVariant.isEmpty()
- && codonVariant.get(0).variant == null)
- {
- /*
- * already recorded base value, add this variant
- */
- codonVariant.get(0).variant = sf;
- }
- else
- {
- /*
- * add variant with base value
- */
- codonVariant.add(new DnaVariant(nucleotide, sf));
- }
+ /*
+ * already recorded base value, add this variant
+ */
+ codonVariant.get(0).variant = sf;
}
- else if (codonVariant.isEmpty())
+ else
{
/*
- * record (possibly non-varying) base value
+ * add variant with base value
*/
- codonVariant.add(new DnaVariant(nucleotide));
+ codonVariant.add(new DnaVariant(nucleotide, sf));
}
}
+ else if (codonVariant.isEmpty())
+ {
+ /*
+ * record (possibly non-varying) base value
+ */
+ codonVariant.add(new DnaVariant(nucleotide));
+ }
}
}
return variants;
seqMap.getMap().getInverse());
}
- char[] fromChars = fromSeq.getSequence();
int toStart = seq.getStart();
- char[] toChars = seq.getSequence();
/*
* traverse [start, end, start, end...] ranges in fromSeq
* of the next character of the mapped-to sequence; stop when all
* the characters of the range have been counted
*/
- while (mappedCharPos <= range[1] && fromCol <= fromChars.length
+ while (mappedCharPos <= range[1] && fromCol <= fromSeq.getLength()
&& fromCol >= 0)
{
- if (!Comparison.isGap(fromChars[fromCol - 1]))
+ if (!Comparison.isGap(fromSeq.getCharAt(fromCol - 1)))
{
/*
* mapped from sequence has a character in this column
seqsMap = new HashMap<SequenceI, Character>();
map.put(fromCol, seqsMap);
}
- seqsMap.put(seq, toChars[mappedCharPos - toStart]);
+ seqsMap.put(seq, seq.getCharAt(mappedCharPos - toStart));
mappedCharPos++;
}
fromCol += (forward ? 1 : -1);